SPOOFGUARD: UNVEILING DECEPTIVE ONLINE PLATFORMS

Artificial Intelligence

Phishing is a cyber-attack used by hackers and malicious organizations to steal the credentials like username, passwords, financial data, and bank details of individuals. They achieve this by pretending to be a legitimate service provider of a popular brand. Phishers create a fake website or clone of a trending website. They try to make the URL of the website like the legitimate one. The user might not notice the minute differences in the website and become the victim of this attack. Though phishing attacks are happening over a long period of time, it is still active and successful. The reason is lack of awareness among the people about the phishing attack. We propose a method that demonstrates the effectiveness of using PSO-based feature weighting to improve the detection of phishing websites and develop that into a Chrome extension which warns the user when the user enters a malicious website. This Chrome extension is designed to enhance users' online safety by identifying and flagging potential phishing websites in real-time. Leveraging advanced machine learning algorithms and PSO, it analyzes various features of web pages, including URL structure, content, and behaviour, to assess the likelihood of a webpage being a phishing site. PSO optimizes the machine learning model's parameters, enhancing its accuracy and adaptability. By incorporating PSO, this extension provides users with instant warnings and prompts them to exercise caution when encountering suspicious websites. This optimization technique allows the machine learning model to adapt and learn from evolving phishing tactics, ensuring its effectiveness against zero- hour attacks. Moreover, its lightweight and user-friendly design seamlessly integrates into users' browsing experiences without causing disruptions. This research contributes to the ongoing efforts to enhance online security and protect users from falling victim to phishing scams.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.